Automated Flowsheet Synthesis Using Hierarchical Reinforcement Learning: Proof of Concept

Recently we showed that reinforcement learning can be used to automatically generate process flowsheets without heuristics or prior knowledge. For this purpose, SynGameZero, a novel two‐player game has been developed. In this work we extend SynGameZero by structuring the agent's actions in several hierarchy levels, which improves the approach in terms of scalability and allows the consideration of more sophisticated flowsheet problems. We successfully demonstrate the usability of our novel framework for the fully automated synthesis of an ethyl tert‐butyl ether process. This work extends a reinforcement learning based approach for automated flowsheet synthesis. An agent is taught to set up flowsheets without prior knowledge by transforming the problem into a two‐player situation. The available actions are split up into several hierarchy levels to improve the original method in terms of scalability.